{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 100.0,
  "global_step": 3500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.29,
      "learning_rate": 4.0000000000000003e-07,
      "loss": 12.428,
      "step": 10
    },
    {
      "epoch": 0.57,
      "learning_rate": 8.500000000000001e-07,
      "loss": 14.8943,
      "step": 20
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.35e-06,
      "loss": 12.502,
      "step": 30
    },
    {
      "epoch": 1.14,
      "learning_rate": 1.85e-06,
      "loss": 13.6247,
      "step": 40
    },
    {
      "epoch": 1.43,
      "learning_rate": 2.35e-06,
      "loss": 13.556,
      "step": 50
    },
    {
      "epoch": 1.71,
      "learning_rate": 2.8500000000000002e-06,
      "loss": 13.0632,
      "step": 60
    },
    {
      "epoch": 2.0,
      "learning_rate": 3.3500000000000005e-06,
      "loss": 13.6556,
      "step": 70
    },
    {
      "epoch": 2.29,
      "learning_rate": 3.85e-06,
      "loss": 12.3356,
      "step": 80
    },
    {
      "epoch": 2.57,
      "learning_rate": 4.35e-06,
      "loss": 14.2951,
      "step": 90
    },
    {
      "epoch": 2.86,
      "learning_rate": 4.85e-06,
      "loss": 12.3533,
      "step": 100
    },
    {
      "epoch": 3.14,
      "learning_rate": 5.3500000000000004e-06,
      "loss": 13.373,
      "step": 110
    },
    {
      "epoch": 3.43,
      "learning_rate": 5.8e-06,
      "loss": 12.7992,
      "step": 120
    },
    {
      "epoch": 3.71,
      "learning_rate": 6.300000000000001e-06,
      "loss": 13.2883,
      "step": 130
    },
    {
      "epoch": 4.0,
      "learning_rate": 6.800000000000001e-06,
      "loss": 12.3322,
      "step": 140
    },
    {
      "epoch": 4.29,
      "learning_rate": 7.2999999999999996e-06,
      "loss": 11.0312,
      "step": 150
    },
    {
      "epoch": 4.57,
      "learning_rate": 7.8e-06,
      "loss": 12.1417,
      "step": 160
    },
    {
      "epoch": 4.86,
      "learning_rate": 8.3e-06,
      "loss": 10.9265,
      "step": 170
    },
    {
      "epoch": 5.14,
      "learning_rate": 8.8e-06,
      "loss": 11.5465,
      "step": 180
    },
    {
      "epoch": 5.43,
      "learning_rate": 9.3e-06,
      "loss": 10.9661,
      "step": 190
    },
    {
      "epoch": 5.71,
      "learning_rate": 9.800000000000001e-06,
      "loss": 11.5904,
      "step": 200
    },
    {
      "epoch": 6.0,
      "learning_rate": 1.03e-05,
      "loss": 11.0434,
      "step": 210
    },
    {
      "epoch": 6.29,
      "learning_rate": 1.08e-05,
      "loss": 10.4085,
      "step": 220
    },
    {
      "epoch": 6.57,
      "learning_rate": 1.13e-05,
      "loss": 11.747,
      "step": 230
    },
    {
      "epoch": 6.86,
      "learning_rate": 1.18e-05,
      "loss": 10.0402,
      "step": 240
    },
    {
      "epoch": 7.14,
      "learning_rate": 1.23e-05,
      "loss": 10.6621,
      "step": 250
    },
    {
      "epoch": 7.43,
      "learning_rate": 1.2800000000000001e-05,
      "loss": 10.2827,
      "step": 260
    },
    {
      "epoch": 7.71,
      "learning_rate": 1.3300000000000001e-05,
      "loss": 11.6184,
      "step": 270
    },
    {
      "epoch": 8.0,
      "learning_rate": 1.3800000000000002e-05,
      "loss": 10.4179,
      "step": 280
    },
    {
      "epoch": 8.29,
      "learning_rate": 1.43e-05,
      "loss": 10.0271,
      "step": 290
    },
    {
      "epoch": 8.57,
      "learning_rate": 1.48e-05,
      "loss": 10.6904,
      "step": 300
    },
    {
      "epoch": 8.86,
      "learning_rate": 1.53e-05,
      "loss": 9.9251,
      "step": 310
    },
    {
      "epoch": 9.14,
      "learning_rate": 1.58e-05,
      "loss": 11.1557,
      "step": 320
    },
    {
      "epoch": 9.43,
      "learning_rate": 1.63e-05,
      "loss": 9.857,
      "step": 330
    },
    {
      "epoch": 9.71,
      "learning_rate": 1.6800000000000002e-05,
      "loss": 10.024,
      "step": 340
    },
    {
      "epoch": 10.0,
      "learning_rate": 1.73e-05,
      "loss": 10.4531,
      "step": 350
    },
    {
      "epoch": 10.29,
      "learning_rate": 1.78e-05,
      "loss": 9.1678,
      "step": 360
    },
    {
      "epoch": 10.57,
      "learning_rate": 1.83e-05,
      "loss": 10.4488,
      "step": 370
    },
    {
      "epoch": 10.86,
      "learning_rate": 1.88e-05,
      "loss": 9.3096,
      "step": 380
    },
    {
      "epoch": 11.14,
      "learning_rate": 1.93e-05,
      "loss": 9.9858,
      "step": 390
    },
    {
      "epoch": 11.43,
      "learning_rate": 1.9800000000000004e-05,
      "loss": 9.304,
      "step": 400
    },
    {
      "epoch": 11.71,
      "learning_rate": 2.0300000000000002e-05,
      "loss": 10.0784,
      "step": 410
    },
    {
      "epoch": 12.0,
      "learning_rate": 2.08e-05,
      "loss": 9.1573,
      "step": 420
    },
    {
      "epoch": 12.29,
      "learning_rate": 2.13e-05,
      "loss": 8.6418,
      "step": 430
    },
    {
      "epoch": 12.57,
      "learning_rate": 2.18e-05,
      "loss": 9.7471,
      "step": 440
    },
    {
      "epoch": 12.86,
      "learning_rate": 2.23e-05,
      "loss": 8.4839,
      "step": 450
    },
    {
      "epoch": 13.14,
      "learning_rate": 2.2800000000000002e-05,
      "loss": 9.2901,
      "step": 460
    },
    {
      "epoch": 13.43,
      "learning_rate": 2.3300000000000004e-05,
      "loss": 8.7123,
      "step": 470
    },
    {
      "epoch": 13.71,
      "learning_rate": 2.38e-05,
      "loss": 8.9374,
      "step": 480
    },
    {
      "epoch": 14.0,
      "learning_rate": 2.43e-05,
      "loss": 8.5236,
      "step": 490
    },
    {
      "epoch": 14.29,
      "learning_rate": 2.48e-05,
      "loss": 8.2135,
      "step": 500
    },
    {
      "epoch": 14.29,
      "eval_loss": 8.425774574279785,
      "eval_runtime": 68.0021,
      "eval_samples_per_second": 49.91,
      "eval_steps_per_second": 0.794,
      "eval_wer": 1.0,
      "step": 500
    },
    {
      "epoch": 14.57,
      "learning_rate": 2.5300000000000002e-05,
      "loss": 8.7061,
      "step": 510
    },
    {
      "epoch": 14.86,
      "learning_rate": 2.58e-05,
      "loss": 7.5734,
      "step": 520
    },
    {
      "epoch": 15.14,
      "learning_rate": 2.6300000000000002e-05,
      "loss": 8.1456,
      "step": 530
    },
    {
      "epoch": 15.43,
      "learning_rate": 2.6800000000000004e-05,
      "loss": 8.0294,
      "step": 540
    },
    {
      "epoch": 15.71,
      "learning_rate": 2.7300000000000003e-05,
      "loss": 7.7208,
      "step": 550
    },
    {
      "epoch": 16.0,
      "learning_rate": 2.7800000000000005e-05,
      "loss": 7.4887,
      "step": 560
    },
    {
      "epoch": 16.29,
      "learning_rate": 2.83e-05,
      "loss": 7.407,
      "step": 570
    },
    {
      "epoch": 16.57,
      "learning_rate": 2.88e-05,
      "loss": 7.6668,
      "step": 580
    },
    {
      "epoch": 16.86,
      "learning_rate": 2.93e-05,
      "loss": 6.5848,
      "step": 590
    },
    {
      "epoch": 17.14,
      "learning_rate": 2.98e-05,
      "loss": 6.8816,
      "step": 600
    },
    {
      "epoch": 17.43,
      "learning_rate": 3.03e-05,
      "loss": 6.6686,
      "step": 610
    },
    {
      "epoch": 17.71,
      "learning_rate": 3.08e-05,
      "loss": 6.7147,
      "step": 620
    },
    {
      "epoch": 18.0,
      "learning_rate": 3.13e-05,
      "loss": 6.5893,
      "step": 630
    },
    {
      "epoch": 18.29,
      "learning_rate": 3.18e-05,
      "loss": 6.1672,
      "step": 640
    },
    {
      "epoch": 18.57,
      "learning_rate": 3.2300000000000006e-05,
      "loss": 6.3244,
      "step": 650
    },
    {
      "epoch": 18.86,
      "learning_rate": 3.2800000000000004e-05,
      "loss": 5.7215,
      "step": 660
    },
    {
      "epoch": 19.14,
      "learning_rate": 3.33e-05,
      "loss": 5.9326,
      "step": 670
    },
    {
      "epoch": 19.43,
      "learning_rate": 3.38e-05,
      "loss": 5.5494,
      "step": 680
    },
    {
      "epoch": 19.71,
      "learning_rate": 3.430000000000001e-05,
      "loss": 5.7156,
      "step": 690
    },
    {
      "epoch": 20.0,
      "learning_rate": 3.48e-05,
      "loss": 5.4507,
      "step": 700
    },
    {
      "epoch": 20.29,
      "learning_rate": 3.53e-05,
      "loss": 4.988,
      "step": 710
    },
    {
      "epoch": 20.57,
      "learning_rate": 3.58e-05,
      "loss": 4.9823,
      "step": 720
    },
    {
      "epoch": 20.86,
      "learning_rate": 3.63e-05,
      "loss": 4.9272,
      "step": 730
    },
    {
      "epoch": 21.14,
      "learning_rate": 3.68e-05,
      "loss": 4.6647,
      "step": 740
    },
    {
      "epoch": 21.43,
      "learning_rate": 3.73e-05,
      "loss": 4.5182,
      "step": 750
    },
    {
      "epoch": 21.71,
      "learning_rate": 3.7800000000000004e-05,
      "loss": 4.3969,
      "step": 760
    },
    {
      "epoch": 22.0,
      "learning_rate": 3.83e-05,
      "loss": 4.32,
      "step": 770
    },
    {
      "epoch": 22.29,
      "learning_rate": 3.88e-05,
      "loss": 3.9023,
      "step": 780
    },
    {
      "epoch": 22.57,
      "learning_rate": 3.9300000000000007e-05,
      "loss": 4.0833,
      "step": 790
    },
    {
      "epoch": 22.86,
      "learning_rate": 3.9800000000000005e-05,
      "loss": 3.8068,
      "step": 800
    },
    {
      "epoch": 23.14,
      "learning_rate": 4.0300000000000004e-05,
      "loss": 3.8671,
      "step": 810
    },
    {
      "epoch": 23.43,
      "learning_rate": 4.08e-05,
      "loss": 3.6491,
      "step": 820
    },
    {
      "epoch": 23.71,
      "learning_rate": 4.13e-05,
      "loss": 3.5627,
      "step": 830
    },
    {
      "epoch": 24.0,
      "learning_rate": 4.18e-05,
      "loss": 3.4637,
      "step": 840
    },
    {
      "epoch": 24.29,
      "learning_rate": 4.23e-05,
      "loss": 3.3317,
      "step": 850
    },
    {
      "epoch": 24.57,
      "learning_rate": 4.2800000000000004e-05,
      "loss": 3.3283,
      "step": 860
    },
    {
      "epoch": 24.86,
      "learning_rate": 4.33e-05,
      "loss": 3.2532,
      "step": 870
    },
    {
      "epoch": 25.14,
      "learning_rate": 4.38e-05,
      "loss": 3.1851,
      "step": 880
    },
    {
      "epoch": 25.43,
      "learning_rate": 4.43e-05,
      "loss": 3.1297,
      "step": 890
    },
    {
      "epoch": 25.71,
      "learning_rate": 4.4800000000000005e-05,
      "loss": 3.1534,
      "step": 900
    },
    {
      "epoch": 26.0,
      "learning_rate": 4.53e-05,
      "loss": 3.1104,
      "step": 910
    },
    {
      "epoch": 26.29,
      "learning_rate": 4.58e-05,
      "loss": 3.0516,
      "step": 920
    },
    {
      "epoch": 26.57,
      "learning_rate": 4.630000000000001e-05,
      "loss": 3.0747,
      "step": 930
    },
    {
      "epoch": 26.86,
      "learning_rate": 4.6800000000000006e-05,
      "loss": 3.0215,
      "step": 940
    },
    {
      "epoch": 27.14,
      "learning_rate": 4.73e-05,
      "loss": 3.053,
      "step": 950
    },
    {
      "epoch": 27.43,
      "learning_rate": 4.78e-05,
      "loss": 3.0019,
      "step": 960
    },
    {
      "epoch": 27.71,
      "learning_rate": 4.83e-05,
      "loss": 3.0177,
      "step": 970
    },
    {
      "epoch": 28.0,
      "learning_rate": 4.88e-05,
      "loss": 3.0276,
      "step": 980
    },
    {
      "epoch": 28.29,
      "learning_rate": 4.93e-05,
      "loss": 3.0244,
      "step": 990
    },
    {
      "epoch": 28.57,
      "learning_rate": 4.9800000000000004e-05,
      "loss": 3.0031,
      "step": 1000
    },
    {
      "epoch": 28.57,
      "eval_loss": 2.983858585357666,
      "eval_runtime": 68.6766,
      "eval_samples_per_second": 49.42,
      "eval_steps_per_second": 0.786,
      "eval_wer": 1.0,
      "step": 1000
    },
    {
      "epoch": 28.86,
      "learning_rate": 4.9880000000000004e-05,
      "loss": 2.9678,
      "step": 1010
    },
    {
      "epoch": 29.14,
      "learning_rate": 4.9680000000000005e-05,
      "loss": 2.997,
      "step": 1020
    },
    {
      "epoch": 29.43,
      "learning_rate": 4.948000000000001e-05,
      "loss": 3.0013,
      "step": 1030
    },
    {
      "epoch": 29.71,
      "learning_rate": 4.928e-05,
      "loss": 2.9827,
      "step": 1040
    },
    {
      "epoch": 30.0,
      "learning_rate": 4.9080000000000004e-05,
      "loss": 3.0162,
      "step": 1050
    },
    {
      "epoch": 30.29,
      "learning_rate": 4.8880000000000006e-05,
      "loss": 2.987,
      "step": 1060
    },
    {
      "epoch": 30.57,
      "learning_rate": 4.868e-05,
      "loss": 2.978,
      "step": 1070
    },
    {
      "epoch": 30.86,
      "learning_rate": 4.8480000000000003e-05,
      "loss": 2.9715,
      "step": 1080
    },
    {
      "epoch": 31.14,
      "learning_rate": 4.8280000000000005e-05,
      "loss": 2.9773,
      "step": 1090
    },
    {
      "epoch": 31.43,
      "learning_rate": 4.808e-05,
      "loss": 2.9853,
      "step": 1100
    },
    {
      "epoch": 31.71,
      "learning_rate": 4.788e-05,
      "loss": 3.0167,
      "step": 1110
    },
    {
      "epoch": 32.0,
      "learning_rate": 4.7680000000000004e-05,
      "loss": 2.9582,
      "step": 1120
    },
    {
      "epoch": 32.29,
      "learning_rate": 4.748e-05,
      "loss": 2.9653,
      "step": 1130
    },
    {
      "epoch": 32.57,
      "learning_rate": 4.728e-05,
      "loss": 3.0065,
      "step": 1140
    },
    {
      "epoch": 32.86,
      "learning_rate": 4.708e-05,
      "loss": 2.9496,
      "step": 1150
    },
    {
      "epoch": 33.14,
      "learning_rate": 4.688e-05,
      "loss": 2.9846,
      "step": 1160
    },
    {
      "epoch": 33.43,
      "learning_rate": 4.668e-05,
      "loss": 2.9646,
      "step": 1170
    },
    {
      "epoch": 33.71,
      "learning_rate": 4.648e-05,
      "loss": 2.9643,
      "step": 1180
    },
    {
      "epoch": 34.0,
      "learning_rate": 4.6280000000000004e-05,
      "loss": 2.9947,
      "step": 1190
    },
    {
      "epoch": 34.29,
      "learning_rate": 4.608e-05,
      "loss": 2.9392,
      "step": 1200
    },
    {
      "epoch": 34.57,
      "learning_rate": 4.588e-05,
      "loss": 2.9746,
      "step": 1210
    },
    {
      "epoch": 34.86,
      "learning_rate": 4.568e-05,
      "loss": 2.9701,
      "step": 1220
    },
    {
      "epoch": 35.14,
      "learning_rate": 4.548e-05,
      "loss": 2.9833,
      "step": 1230
    },
    {
      "epoch": 35.43,
      "learning_rate": 4.528e-05,
      "loss": 2.9887,
      "step": 1240
    },
    {
      "epoch": 35.71,
      "learning_rate": 4.508e-05,
      "loss": 2.9917,
      "step": 1250
    },
    {
      "epoch": 36.0,
      "learning_rate": 4.488e-05,
      "loss": 2.9381,
      "step": 1260
    },
    {
      "epoch": 36.29,
      "learning_rate": 4.468e-05,
      "loss": 2.9206,
      "step": 1270
    },
    {
      "epoch": 36.57,
      "learning_rate": 4.448e-05,
      "loss": 3.0081,
      "step": 1280
    },
    {
      "epoch": 36.86,
      "learning_rate": 4.428e-05,
      "loss": 2.9628,
      "step": 1290
    },
    {
      "epoch": 37.14,
      "learning_rate": 4.4080000000000005e-05,
      "loss": 2.9533,
      "step": 1300
    },
    {
      "epoch": 37.43,
      "learning_rate": 4.388000000000001e-05,
      "loss": 2.9469,
      "step": 1310
    },
    {
      "epoch": 37.71,
      "learning_rate": 4.368e-05,
      "loss": 2.9682,
      "step": 1320
    },
    {
      "epoch": 38.0,
      "learning_rate": 4.3480000000000004e-05,
      "loss": 2.9994,
      "step": 1330
    },
    {
      "epoch": 38.29,
      "learning_rate": 4.3280000000000006e-05,
      "loss": 2.969,
      "step": 1340
    },
    {
      "epoch": 38.57,
      "learning_rate": 4.308e-05,
      "loss": 2.9899,
      "step": 1350
    },
    {
      "epoch": 38.86,
      "learning_rate": 4.288e-05,
      "loss": 2.9271,
      "step": 1360
    },
    {
      "epoch": 39.14,
      "learning_rate": 4.2680000000000005e-05,
      "loss": 2.9551,
      "step": 1370
    },
    {
      "epoch": 39.43,
      "learning_rate": 4.248e-05,
      "loss": 2.9454,
      "step": 1380
    },
    {
      "epoch": 39.71,
      "learning_rate": 4.228e-05,
      "loss": 2.9995,
      "step": 1390
    },
    {
      "epoch": 40.0,
      "learning_rate": 4.2080000000000004e-05,
      "loss": 2.9422,
      "step": 1400
    },
    {
      "epoch": 40.29,
      "learning_rate": 4.1880000000000006e-05,
      "loss": 2.9308,
      "step": 1410
    },
    {
      "epoch": 40.57,
      "learning_rate": 4.168e-05,
      "loss": 3.0313,
      "step": 1420
    },
    {
      "epoch": 40.86,
      "learning_rate": 4.148e-05,
      "loss": 2.9375,
      "step": 1430
    },
    {
      "epoch": 41.14,
      "learning_rate": 4.1280000000000005e-05,
      "loss": 2.9452,
      "step": 1440
    },
    {
      "epoch": 41.43,
      "learning_rate": 4.108e-05,
      "loss": 2.9515,
      "step": 1450
    },
    {
      "epoch": 41.71,
      "learning_rate": 4.088e-05,
      "loss": 2.9442,
      "step": 1460
    },
    {
      "epoch": 42.0,
      "learning_rate": 4.0680000000000004e-05,
      "loss": 2.9973,
      "step": 1470
    },
    {
      "epoch": 42.29,
      "learning_rate": 4.048e-05,
      "loss": 2.9598,
      "step": 1480
    },
    {
      "epoch": 42.57,
      "learning_rate": 4.028e-05,
      "loss": 2.9512,
      "step": 1490
    },
    {
      "epoch": 42.86,
      "learning_rate": 4.008e-05,
      "loss": 2.9661,
      "step": 1500
    },
    {
      "epoch": 42.86,
      "eval_loss": 2.9401793479919434,
      "eval_runtime": 67.8098,
      "eval_samples_per_second": 50.052,
      "eval_steps_per_second": 0.796,
      "eval_wer": 1.0,
      "step": 1500
    },
    {
      "epoch": 43.14,
      "learning_rate": 3.988e-05,
      "loss": 2.9636,
      "step": 1510
    },
    {
      "epoch": 43.43,
      "learning_rate": 3.968e-05,
      "loss": 2.9522,
      "step": 1520
    },
    {
      "epoch": 43.71,
      "learning_rate": 3.948e-05,
      "loss": 2.9481,
      "step": 1530
    },
    {
      "epoch": 44.0,
      "learning_rate": 3.9280000000000003e-05,
      "loss": 2.9753,
      "step": 1540
    },
    {
      "epoch": 44.29,
      "learning_rate": 3.908e-05,
      "loss": 2.9865,
      "step": 1550
    },
    {
      "epoch": 44.57,
      "learning_rate": 3.888e-05,
      "loss": 2.9511,
      "step": 1560
    },
    {
      "epoch": 44.86,
      "learning_rate": 3.868e-05,
      "loss": 2.9181,
      "step": 1570
    },
    {
      "epoch": 45.14,
      "learning_rate": 3.848e-05,
      "loss": 2.9562,
      "step": 1580
    },
    {
      "epoch": 45.43,
      "learning_rate": 3.828e-05,
      "loss": 2.9485,
      "step": 1590
    },
    {
      "epoch": 45.71,
      "learning_rate": 3.808e-05,
      "loss": 2.989,
      "step": 1600
    },
    {
      "epoch": 46.0,
      "learning_rate": 3.788e-05,
      "loss": 2.9678,
      "step": 1610
    },
    {
      "epoch": 46.29,
      "learning_rate": 3.7680000000000005e-05,
      "loss": 2.9232,
      "step": 1620
    },
    {
      "epoch": 46.57,
      "learning_rate": 3.748000000000001e-05,
      "loss": 3.0069,
      "step": 1630
    },
    {
      "epoch": 46.86,
      "learning_rate": 3.728e-05,
      "loss": 2.9221,
      "step": 1640
    },
    {
      "epoch": 47.14,
      "learning_rate": 3.7080000000000004e-05,
      "loss": 2.9606,
      "step": 1650
    },
    {
      "epoch": 47.43,
      "learning_rate": 3.6880000000000006e-05,
      "loss": 2.9592,
      "step": 1660
    },
    {
      "epoch": 47.71,
      "learning_rate": 3.668e-05,
      "loss": 2.9769,
      "step": 1670
    },
    {
      "epoch": 48.0,
      "learning_rate": 3.648e-05,
      "loss": 2.9492,
      "step": 1680
    },
    {
      "epoch": 48.29,
      "learning_rate": 3.6280000000000005e-05,
      "loss": 2.9682,
      "step": 1690
    },
    {
      "epoch": 48.57,
      "learning_rate": 3.608e-05,
      "loss": 2.9818,
      "step": 1700
    },
    {
      "epoch": 48.86,
      "learning_rate": 3.588e-05,
      "loss": 2.9231,
      "step": 1710
    },
    {
      "epoch": 49.14,
      "learning_rate": 3.5680000000000004e-05,
      "loss": 2.9391,
      "step": 1720
    },
    {
      "epoch": 49.43,
      "learning_rate": 3.548e-05,
      "loss": 2.9502,
      "step": 1730
    },
    {
      "epoch": 49.71,
      "learning_rate": 3.528e-05,
      "loss": 2.9398,
      "step": 1740
    },
    {
      "epoch": 50.0,
      "learning_rate": 3.508e-05,
      "loss": 3.0095,
      "step": 1750
    },
    {
      "epoch": 50.29,
      "learning_rate": 3.4880000000000005e-05,
      "loss": 2.9326,
      "step": 1760
    },
    {
      "epoch": 50.57,
      "learning_rate": 3.468e-05,
      "loss": 2.9808,
      "step": 1770
    },
    {
      "epoch": 50.86,
      "learning_rate": 3.448e-05,
      "loss": 2.9393,
      "step": 1780
    },
    {
      "epoch": 51.14,
      "learning_rate": 3.4280000000000004e-05,
      "loss": 2.9855,
      "step": 1790
    },
    {
      "epoch": 51.43,
      "learning_rate": 3.408e-05,
      "loss": 2.9565,
      "step": 1800
    },
    {
      "epoch": 51.71,
      "learning_rate": 3.388e-05,
      "loss": 2.9555,
      "step": 1810
    },
    {
      "epoch": 52.0,
      "learning_rate": 3.368e-05,
      "loss": 2.9553,
      "step": 1820
    },
    {
      "epoch": 52.29,
      "learning_rate": 3.348e-05,
      "loss": 2.9403,
      "step": 1830
    },
    {
      "epoch": 52.57,
      "learning_rate": 3.328e-05,
      "loss": 2.9752,
      "step": 1840
    },
    {
      "epoch": 52.86,
      "learning_rate": 3.308e-05,
      "loss": 2.9453,
      "step": 1850
    },
    {
      "epoch": 53.14,
      "learning_rate": 3.288e-05,
      "loss": 2.9319,
      "step": 1860
    },
    {
      "epoch": 53.43,
      "learning_rate": 3.268e-05,
      "loss": 2.9418,
      "step": 1870
    },
    {
      "epoch": 53.71,
      "learning_rate": 3.248e-05,
      "loss": 2.9767,
      "step": 1880
    },
    {
      "epoch": 54.0,
      "learning_rate": 3.2279999999999996e-05,
      "loss": 2.982,
      "step": 1890
    },
    {
      "epoch": 54.29,
      "learning_rate": 3.208e-05,
      "loss": 2.9387,
      "step": 1900
    },
    {
      "epoch": 54.57,
      "learning_rate": 3.188e-05,
      "loss": 2.9649,
      "step": 1910
    },
    {
      "epoch": 54.86,
      "learning_rate": 3.168e-05,
      "loss": 2.9581,
      "step": 1920
    },
    {
      "epoch": 55.14,
      "learning_rate": 3.1480000000000004e-05,
      "loss": 2.9187,
      "step": 1930
    },
    {
      "epoch": 55.43,
      "learning_rate": 3.1280000000000005e-05,
      "loss": 2.9783,
      "step": 1940
    },
    {
      "epoch": 55.71,
      "learning_rate": 3.108e-05,
      "loss": 2.9672,
      "step": 1950
    },
    {
      "epoch": 56.0,
      "learning_rate": 3.088e-05,
      "loss": 2.9568,
      "step": 1960
    },
    {
      "epoch": 56.29,
      "learning_rate": 3.0680000000000004e-05,
      "loss": 2.9539,
      "step": 1970
    },
    {
      "epoch": 56.57,
      "learning_rate": 3.0480000000000003e-05,
      "loss": 2.9583,
      "step": 1980
    },
    {
      "epoch": 56.86,
      "learning_rate": 3.028e-05,
      "loss": 2.9304,
      "step": 1990
    },
    {
      "epoch": 57.14,
      "learning_rate": 3.0080000000000003e-05,
      "loss": 2.9584,
      "step": 2000
    },
    {
      "epoch": 57.14,
      "eval_loss": 2.935370683670044,
      "eval_runtime": 68.0892,
      "eval_samples_per_second": 49.846,
      "eval_steps_per_second": 0.793,
      "eval_wer": 1.0,
      "step": 2000
    },
    {
      "epoch": 57.43,
      "learning_rate": 2.9880000000000002e-05,
      "loss": 2.9527,
      "step": 2010
    },
    {
      "epoch": 57.71,
      "learning_rate": 2.9680000000000004e-05,
      "loss": 2.9761,
      "step": 2020
    },
    {
      "epoch": 58.0,
      "learning_rate": 2.9480000000000002e-05,
      "loss": 2.9374,
      "step": 2030
    },
    {
      "epoch": 58.29,
      "learning_rate": 2.928e-05,
      "loss": 2.9363,
      "step": 2040
    },
    {
      "epoch": 58.57,
      "learning_rate": 2.9080000000000003e-05,
      "loss": 2.9542,
      "step": 2050
    },
    {
      "epoch": 58.86,
      "learning_rate": 2.888e-05,
      "loss": 2.9437,
      "step": 2060
    },
    {
      "epoch": 59.14,
      "learning_rate": 2.868e-05,
      "loss": 2.9649,
      "step": 2070
    },
    {
      "epoch": 59.43,
      "learning_rate": 2.8480000000000002e-05,
      "loss": 2.9728,
      "step": 2080
    },
    {
      "epoch": 59.71,
      "learning_rate": 2.828e-05,
      "loss": 2.9504,
      "step": 2090
    },
    {
      "epoch": 60.0,
      "learning_rate": 2.8080000000000002e-05,
      "loss": 2.9564,
      "step": 2100
    },
    {
      "epoch": 60.29,
      "learning_rate": 2.788e-05,
      "loss": 2.9248,
      "step": 2110
    },
    {
      "epoch": 60.57,
      "learning_rate": 2.768e-05,
      "loss": 2.9838,
      "step": 2120
    },
    {
      "epoch": 60.86,
      "learning_rate": 2.748e-05,
      "loss": 2.953,
      "step": 2130
    },
    {
      "epoch": 61.14,
      "learning_rate": 2.728e-05,
      "loss": 2.9591,
      "step": 2140
    },
    {
      "epoch": 61.43,
      "learning_rate": 2.7079999999999998e-05,
      "loss": 2.9299,
      "step": 2150
    },
    {
      "epoch": 61.71,
      "learning_rate": 2.688e-05,
      "loss": 2.9672,
      "step": 2160
    },
    {
      "epoch": 62.0,
      "learning_rate": 2.668e-05,
      "loss": 3.0012,
      "step": 2170
    },
    {
      "epoch": 62.29,
      "learning_rate": 2.648e-05,
      "loss": 2.9727,
      "step": 2180
    },
    {
      "epoch": 62.57,
      "learning_rate": 2.628e-05,
      "loss": 2.9434,
      "step": 2190
    },
    {
      "epoch": 62.86,
      "learning_rate": 2.6079999999999998e-05,
      "loss": 2.9317,
      "step": 2200
    },
    {
      "epoch": 63.14,
      "learning_rate": 2.588e-05,
      "loss": 2.9616,
      "step": 2210
    },
    {
      "epoch": 63.43,
      "learning_rate": 2.5679999999999998e-05,
      "loss": 2.9577,
      "step": 2220
    },
    {
      "epoch": 63.71,
      "learning_rate": 2.5480000000000003e-05,
      "loss": 2.9803,
      "step": 2230
    },
    {
      "epoch": 64.0,
      "learning_rate": 2.5280000000000005e-05,
      "loss": 2.9413,
      "step": 2240
    },
    {
      "epoch": 64.29,
      "learning_rate": 2.5080000000000004e-05,
      "loss": 2.926,
      "step": 2250
    },
    {
      "epoch": 64.57,
      "learning_rate": 2.488e-05,
      "loss": 2.9775,
      "step": 2260
    },
    {
      "epoch": 64.86,
      "learning_rate": 2.468e-05,
      "loss": 2.9351,
      "step": 2270
    },
    {
      "epoch": 65.14,
      "learning_rate": 2.448e-05,
      "loss": 2.9719,
      "step": 2280
    },
    {
      "epoch": 65.43,
      "learning_rate": 2.428e-05,
      "loss": 2.965,
      "step": 2290
    },
    {
      "epoch": 65.71,
      "learning_rate": 2.408e-05,
      "loss": 2.9687,
      "step": 2300
    },
    {
      "epoch": 66.0,
      "learning_rate": 2.3880000000000002e-05,
      "loss": 2.9313,
      "step": 2310
    },
    {
      "epoch": 66.29,
      "learning_rate": 2.3680000000000004e-05,
      "loss": 2.9569,
      "step": 2320
    },
    {
      "epoch": 66.57,
      "learning_rate": 2.3480000000000002e-05,
      "loss": 2.9427,
      "step": 2330
    },
    {
      "epoch": 66.86,
      "learning_rate": 2.328e-05,
      "loss": 2.945,
      "step": 2340
    },
    {
      "epoch": 67.14,
      "learning_rate": 2.3080000000000003e-05,
      "loss": 2.948,
      "step": 2350
    },
    {
      "epoch": 67.43,
      "learning_rate": 2.288e-05,
      "loss": 2.953,
      "step": 2360
    },
    {
      "epoch": 67.71,
      "learning_rate": 2.268e-05,
      "loss": 2.9935,
      "step": 2370
    },
    {
      "epoch": 68.0,
      "learning_rate": 2.248e-05,
      "loss": 2.9362,
      "step": 2380
    },
    {
      "epoch": 68.29,
      "learning_rate": 2.228e-05,
      "loss": 2.9103,
      "step": 2390
    },
    {
      "epoch": 68.57,
      "learning_rate": 2.2080000000000002e-05,
      "loss": 2.9818,
      "step": 2400
    },
    {
      "epoch": 68.86,
      "learning_rate": 2.188e-05,
      "loss": 2.9291,
      "step": 2410
    },
    {
      "epoch": 69.14,
      "learning_rate": 2.168e-05,
      "loss": 2.991,
      "step": 2420
    },
    {
      "epoch": 69.43,
      "learning_rate": 2.148e-05,
      "loss": 2.9405,
      "step": 2430
    },
    {
      "epoch": 69.71,
      "learning_rate": 2.128e-05,
      "loss": 2.9717,
      "step": 2440
    },
    {
      "epoch": 70.0,
      "learning_rate": 2.1079999999999998e-05,
      "loss": 2.9638,
      "step": 2450
    },
    {
      "epoch": 70.29,
      "learning_rate": 2.0880000000000003e-05,
      "loss": 2.9538,
      "step": 2460
    },
    {
      "epoch": 70.57,
      "learning_rate": 2.0680000000000002e-05,
      "loss": 2.9574,
      "step": 2470
    },
    {
      "epoch": 70.86,
      "learning_rate": 2.048e-05,
      "loss": 2.9297,
      "step": 2480
    },
    {
      "epoch": 71.14,
      "learning_rate": 2.0280000000000002e-05,
      "loss": 2.9352,
      "step": 2490
    },
    {
      "epoch": 71.43,
      "learning_rate": 2.008e-05,
      "loss": 2.936,
      "step": 2500
    },
    {
      "epoch": 71.43,
      "eval_loss": 2.934075355529785,
      "eval_runtime": 68.513,
      "eval_samples_per_second": 49.538,
      "eval_steps_per_second": 0.788,
      "eval_wer": 1.0,
      "step": 2500
    },
    {
      "epoch": 71.71,
      "learning_rate": 1.9880000000000003e-05,
      "loss": 3.0163,
      "step": 2510
    },
    {
      "epoch": 72.0,
      "learning_rate": 1.968e-05,
      "loss": 2.925,
      "step": 2520
    },
    {
      "epoch": 72.29,
      "learning_rate": 1.948e-05,
      "loss": 2.943,
      "step": 2530
    },
    {
      "epoch": 72.57,
      "learning_rate": 1.9280000000000002e-05,
      "loss": 2.9606,
      "step": 2540
    },
    {
      "epoch": 72.86,
      "learning_rate": 1.908e-05,
      "loss": 2.9465,
      "step": 2550
    },
    {
      "epoch": 73.14,
      "learning_rate": 1.888e-05,
      "loss": 2.948,
      "step": 2560
    },
    {
      "epoch": 73.43,
      "learning_rate": 1.868e-05,
      "loss": 2.939,
      "step": 2570
    },
    {
      "epoch": 73.71,
      "learning_rate": 1.848e-05,
      "loss": 2.9423,
      "step": 2580
    },
    {
      "epoch": 74.0,
      "learning_rate": 1.828e-05,
      "loss": 2.9882,
      "step": 2590
    },
    {
      "epoch": 74.29,
      "learning_rate": 1.808e-05,
      "loss": 2.9244,
      "step": 2600
    },
    {
      "epoch": 74.57,
      "learning_rate": 1.7879999999999998e-05,
      "loss": 2.9544,
      "step": 2610
    },
    {
      "epoch": 74.86,
      "learning_rate": 1.7680000000000004e-05,
      "loss": 2.9475,
      "step": 2620
    },
    {
      "epoch": 75.14,
      "learning_rate": 1.7480000000000002e-05,
      "loss": 2.9681,
      "step": 2630
    },
    {
      "epoch": 75.43,
      "learning_rate": 1.728e-05,
      "loss": 2.9591,
      "step": 2640
    },
    {
      "epoch": 75.71,
      "learning_rate": 1.7080000000000002e-05,
      "loss": 2.9402,
      "step": 2650
    },
    {
      "epoch": 76.0,
      "learning_rate": 1.688e-05,
      "loss": 2.9719,
      "step": 2660
    },
    {
      "epoch": 76.29,
      "learning_rate": 1.668e-05,
      "loss": 2.9784,
      "step": 2670
    },
    {
      "epoch": 76.57,
      "learning_rate": 1.648e-05,
      "loss": 2.9333,
      "step": 2680
    },
    {
      "epoch": 76.86,
      "learning_rate": 1.628e-05,
      "loss": 2.934,
      "step": 2690
    },
    {
      "epoch": 77.14,
      "learning_rate": 1.6080000000000002e-05,
      "loss": 2.9378,
      "step": 2700
    },
    {
      "epoch": 77.43,
      "learning_rate": 1.588e-05,
      "loss": 2.9637,
      "step": 2710
    },
    {
      "epoch": 77.71,
      "learning_rate": 1.568e-05,
      "loss": 2.9292,
      "step": 2720
    },
    {
      "epoch": 78.0,
      "learning_rate": 1.548e-05,
      "loss": 2.9873,
      "step": 2730
    },
    {
      "epoch": 78.29,
      "learning_rate": 1.528e-05,
      "loss": 2.9161,
      "step": 2740
    },
    {
      "epoch": 78.57,
      "learning_rate": 1.508e-05,
      "loss": 2.9813,
      "step": 2750
    },
    {
      "epoch": 78.86,
      "learning_rate": 1.488e-05,
      "loss": 2.9379,
      "step": 2760
    },
    {
      "epoch": 79.14,
      "learning_rate": 1.4680000000000002e-05,
      "loss": 2.9491,
      "step": 2770
    },
    {
      "epoch": 79.43,
      "learning_rate": 1.4480000000000002e-05,
      "loss": 2.9589,
      "step": 2780
    },
    {
      "epoch": 79.71,
      "learning_rate": 1.4280000000000002e-05,
      "loss": 2.9501,
      "step": 2790
    },
    {
      "epoch": 80.0,
      "learning_rate": 1.408e-05,
      "loss": 2.9705,
      "step": 2800
    },
    {
      "epoch": 80.29,
      "learning_rate": 1.3880000000000001e-05,
      "loss": 2.9437,
      "step": 2810
    },
    {
      "epoch": 80.57,
      "learning_rate": 1.3680000000000001e-05,
      "loss": 2.9993,
      "step": 2820
    },
    {
      "epoch": 80.86,
      "learning_rate": 1.3480000000000001e-05,
      "loss": 2.9357,
      "step": 2830
    },
    {
      "epoch": 81.14,
      "learning_rate": 1.3280000000000002e-05,
      "loss": 2.913,
      "step": 2840
    },
    {
      "epoch": 81.43,
      "learning_rate": 1.308e-05,
      "loss": 2.9592,
      "step": 2850
    },
    {
      "epoch": 81.71,
      "learning_rate": 1.288e-05,
      "loss": 2.9439,
      "step": 2860
    },
    {
      "epoch": 82.0,
      "learning_rate": 1.268e-05,
      "loss": 2.9795,
      "step": 2870
    },
    {
      "epoch": 82.29,
      "learning_rate": 1.248e-05,
      "loss": 2.9406,
      "step": 2880
    },
    {
      "epoch": 82.57,
      "learning_rate": 1.2280000000000001e-05,
      "loss": 2.9497,
      "step": 2890
    },
    {
      "epoch": 82.86,
      "learning_rate": 1.2080000000000001e-05,
      "loss": 2.9572,
      "step": 2900
    },
    {
      "epoch": 83.14,
      "learning_rate": 1.1880000000000001e-05,
      "loss": 2.9361,
      "step": 2910
    },
    {
      "epoch": 83.43,
      "learning_rate": 1.168e-05,
      "loss": 2.9595,
      "step": 2920
    },
    {
      "epoch": 83.71,
      "learning_rate": 1.148e-05,
      "loss": 2.9364,
      "step": 2930
    },
    {
      "epoch": 84.0,
      "learning_rate": 1.128e-05,
      "loss": 2.9881,
      "step": 2940
    },
    {
      "epoch": 84.29,
      "learning_rate": 1.108e-05,
      "loss": 2.9374,
      "step": 2950
    },
    {
      "epoch": 84.57,
      "learning_rate": 1.088e-05,
      "loss": 2.9347,
      "step": 2960
    },
    {
      "epoch": 84.86,
      "learning_rate": 1.0680000000000001e-05,
      "loss": 2.9809,
      "step": 2970
    },
    {
      "epoch": 85.14,
      "learning_rate": 1.0480000000000001e-05,
      "loss": 2.9295,
      "step": 2980
    },
    {
      "epoch": 85.43,
      "learning_rate": 1.0280000000000002e-05,
      "loss": 2.9674,
      "step": 2990
    },
    {
      "epoch": 85.71,
      "learning_rate": 1.008e-05,
      "loss": 2.9344,
      "step": 3000
    },
    {
      "epoch": 85.71,
      "eval_loss": 2.9323294162750244,
      "eval_runtime": 68.6833,
      "eval_samples_per_second": 49.415,
      "eval_steps_per_second": 0.786,
      "eval_wer": 1.0,
      "step": 3000
    },
    {
      "epoch": 86.0,
      "learning_rate": 9.88e-06,
      "loss": 2.9698,
      "step": 3010
    },
    {
      "epoch": 86.29,
      "learning_rate": 9.68e-06,
      "loss": 2.9697,
      "step": 3020
    },
    {
      "epoch": 86.57,
      "learning_rate": 9.48e-06,
      "loss": 2.9508,
      "step": 3030
    },
    {
      "epoch": 86.86,
      "learning_rate": 9.28e-06,
      "loss": 2.9293,
      "step": 3040
    },
    {
      "epoch": 87.14,
      "learning_rate": 9.080000000000001e-06,
      "loss": 2.9379,
      "step": 3050
    },
    {
      "epoch": 87.43,
      "learning_rate": 8.880000000000001e-06,
      "loss": 2.964,
      "step": 3060
    },
    {
      "epoch": 87.71,
      "learning_rate": 8.68e-06,
      "loss": 2.9583,
      "step": 3070
    },
    {
      "epoch": 88.0,
      "learning_rate": 8.48e-06,
      "loss": 2.9483,
      "step": 3080
    },
    {
      "epoch": 88.29,
      "learning_rate": 8.28e-06,
      "loss": 2.9244,
      "step": 3090
    },
    {
      "epoch": 88.57,
      "learning_rate": 8.08e-06,
      "loss": 2.9881,
      "step": 3100
    },
    {
      "epoch": 88.86,
      "learning_rate": 7.879999999999999e-06,
      "loss": 2.968,
      "step": 3110
    },
    {
      "epoch": 89.14,
      "learning_rate": 7.68e-06,
      "loss": 2.9178,
      "step": 3120
    },
    {
      "epoch": 89.43,
      "learning_rate": 7.480000000000001e-06,
      "loss": 2.9697,
      "step": 3130
    },
    {
      "epoch": 89.71,
      "learning_rate": 7.280000000000001e-06,
      "loss": 2.9473,
      "step": 3140
    },
    {
      "epoch": 90.0,
      "learning_rate": 7.080000000000001e-06,
      "loss": 2.9576,
      "step": 3150
    },
    {
      "epoch": 90.29,
      "learning_rate": 6.88e-06,
      "loss": 2.9285,
      "step": 3160
    },
    {
      "epoch": 90.57,
      "learning_rate": 6.68e-06,
      "loss": 2.9644,
      "step": 3170
    },
    {
      "epoch": 90.86,
      "learning_rate": 6.48e-06,
      "loss": 2.9455,
      "step": 3180
    },
    {
      "epoch": 91.14,
      "learning_rate": 6.28e-06,
      "loss": 2.9929,
      "step": 3190
    },
    {
      "epoch": 91.43,
      "learning_rate": 6.08e-06,
      "loss": 2.9265,
      "step": 3200
    },
    {
      "epoch": 91.71,
      "learning_rate": 5.8800000000000005e-06,
      "loss": 2.9346,
      "step": 3210
    },
    {
      "epoch": 92.0,
      "learning_rate": 5.680000000000001e-06,
      "loss": 2.977,
      "step": 3220
    },
    {
      "epoch": 92.29,
      "learning_rate": 5.48e-06,
      "loss": 2.9082,
      "step": 3230
    },
    {
      "epoch": 92.57,
      "learning_rate": 5.28e-06,
      "loss": 2.9529,
      "step": 3240
    },
    {
      "epoch": 92.86,
      "learning_rate": 5.08e-06,
      "loss": 2.9641,
      "step": 3250
    },
    {
      "epoch": 93.14,
      "learning_rate": 4.880000000000001e-06,
      "loss": 2.9695,
      "step": 3260
    },
    {
      "epoch": 93.43,
      "learning_rate": 4.68e-06,
      "loss": 2.9437,
      "step": 3270
    },
    {
      "epoch": 93.71,
      "learning_rate": 4.48e-06,
      "loss": 2.9812,
      "step": 3280
    },
    {
      "epoch": 94.0,
      "learning_rate": 4.28e-06,
      "loss": 2.9305,
      "step": 3290
    },
    {
      "epoch": 94.29,
      "learning_rate": 4.080000000000001e-06,
      "loss": 2.9409,
      "step": 3300
    },
    {
      "epoch": 94.57,
      "learning_rate": 3.88e-06,
      "loss": 2.9599,
      "step": 3310
    },
    {
      "epoch": 94.86,
      "learning_rate": 3.68e-06,
      "loss": 2.961,
      "step": 3320
    },
    {
      "epoch": 95.14,
      "learning_rate": 3.4799999999999997e-06,
      "loss": 2.9488,
      "step": 3330
    },
    {
      "epoch": 95.43,
      "learning_rate": 3.2800000000000004e-06,
      "loss": 2.917,
      "step": 3340
    },
    {
      "epoch": 95.71,
      "learning_rate": 3.08e-06,
      "loss": 2.9744,
      "step": 3350
    },
    {
      "epoch": 96.0,
      "learning_rate": 2.88e-06,
      "loss": 2.959,
      "step": 3360
    },
    {
      "epoch": 96.29,
      "learning_rate": 2.68e-06,
      "loss": 2.9471,
      "step": 3370
    },
    {
      "epoch": 96.57,
      "learning_rate": 2.48e-06,
      "loss": 2.9788,
      "step": 3380
    },
    {
      "epoch": 96.86,
      "learning_rate": 2.28e-06,
      "loss": 2.9368,
      "step": 3390
    },
    {
      "epoch": 97.14,
      "learning_rate": 2.08e-06,
      "loss": 2.9444,
      "step": 3400
    },
    {
      "epoch": 97.43,
      "learning_rate": 1.8800000000000002e-06,
      "loss": 2.942,
      "step": 3410
    },
    {
      "epoch": 97.71,
      "learning_rate": 1.68e-06,
      "loss": 2.9754,
      "step": 3420
    },
    {
      "epoch": 98.0,
      "learning_rate": 1.4800000000000002e-06,
      "loss": 2.9453,
      "step": 3430
    },
    {
      "epoch": 98.29,
      "learning_rate": 1.28e-06,
      "loss": 2.9342,
      "step": 3440
    },
    {
      "epoch": 98.57,
      "learning_rate": 1.08e-06,
      "loss": 2.9914,
      "step": 3450
    },
    {
      "epoch": 98.86,
      "learning_rate": 8.8e-07,
      "loss": 2.9372,
      "step": 3460
    },
    {
      "epoch": 99.14,
      "learning_rate": 6.8e-07,
      "loss": 2.931,
      "step": 3470
    },
    {
      "epoch": 99.43,
      "learning_rate": 4.8e-07,
      "loss": 2.9347,
      "step": 3480
    },
    {
      "epoch": 99.71,
      "learning_rate": 2.8e-07,
      "loss": 2.9643,
      "step": 3490
    },
    {
      "epoch": 100.0,
      "learning_rate": 8e-08,
      "loss": 2.9674,
      "step": 3500
    },
    {
      "epoch": 100.0,
      "eval_loss": 2.9325411319732666,
      "eval_runtime": 67.927,
      "eval_samples_per_second": 49.965,
      "eval_steps_per_second": 0.795,
      "eval_wer": 1.0,
      "step": 3500
    },
    {
      "epoch": 100.0,
      "step": 3500,
      "total_flos": 3.191123523691086e+19,
      "train_loss": 4.383838545663016,
      "train_runtime": 5188.6933,
      "train_samples_per_second": 42.284,
      "train_steps_per_second": 0.675
    }
  ],
  "max_steps": 3500,
  "num_train_epochs": 100,
  "total_flos": 3.191123523691086e+19,
  "trial_name": null,
  "trial_params": null
}